Efalizumab and etanercept for the treatment of plaque psoriasis
Comments compiled by A.D. Ormerod on behalf of the British Association of
Dermatologists
Executive summary
We wish to submit the British Association of Dermatologists Guidelines for the use of
biological interventions in psoriasis and to highlight discrepancies between the appraisal by
NICE and our own interpretation of the evidence. The views expressed herein are the result
of wide consultation among the membership of the BAD.
We strongly support the aim of the NICE appraisal to provide clear, evidenced based
guidance on use of biological interventions in psoriasis.
There are several key areas where there are clinically important differences between our
guidelines and those in the NICE appraisal, which if implemented would have profound
adverse effects on patients and their clinical care.
►Eligibility criteria and assessment of disease response must include a quality of life
indicator in addition to the PASI score given the nature of psoriasis, and to achieve parity
with NICE guidance in other disease areas
►We question the evidence to support a PASI score of more than 20 for eligibility. This
we feel is inappropriate. Similarly the requirement for a PASI improvement of 75% at 12
weeks will deny treatment to a significant group of patients with a worthwhile response who
go on improving over time. A PASI improvement of 50% at 12 weeks is considered a better
target.
►We feel there will be advantages to having the facility to utilise a higher dose (50mg
twice weekly) of etanercept in some situations where greater efficacy is required.
►Dictating the sequence in which the agents should be used is not appropriate. There is
no evidence of significant differences in efficacy between the two agents and this is further
supported by recent data. The economic arguments given in the appraisal are incomplete and
lack sufficient detail to be persuasive. For example, intermittent treatment should not be
compared to continuous treatment of a chronic disease since a patient relapsing and being
retreated will not have equivalent improvements in quality of life. The BAD guidelines
suggest that a patient centred clinical decision would form the basis of which agent to use
first. We believe that new evidence warrants consideration. We believe that a PASI score
should not be the only consideration and that quality of life should be considered in
eligibility. We question the evidence to support a PASI score of more than 20 for eligibility.
This we feel is inappropriate. Similarly the requirement for a PASI improvement of 75% at
12 weeks will deny treatment to a significant group of patients with a worthwhile response
who go on improving over time. A PASI improvement of 50% at 12 weeks is considered a
1
better target. We feel there will be advantages to having the facility to utilise a higher dose
(50mg twice weekly) of etanercept in situations where greater efficacy is required.
Intermittent treatment should not be compared to continuous treatment of a chronic disease.
A patient relapsing and being retreated will not have equivalent improvements in quality of
life. This needs to be factored into the economic models. There is not evidence of significant
differences in efficacy between the two agents and this is further supported by recent data.
Therefore dictating the sequence in which the agents should be used is not appropriate. The
economic arguments given in the appraisal are incomplete and lack sufficient detail to be
persuasive. The BAD guidelines suggest that a patient centred clinical decision would form
the basis of which agent to use first. Similarly our guidelines give better definition of what
constitutes failure of conventional agents.
1. Whether you consider that all of the relevant evidence has been taken into
account
A working group of the British Association of Dermatologists has just finalised an evidence
based guideline for treatment of psoriasis with rigorous methodology. This was based on
clinical concerns and review of the evidence. We did not have any economic assessment as
part of this process and appreciate that NICE have this highly important additional
information. I will attach the final guidelines for your consideration and will highlight where
the guidelines differ from your draft evaluation.
We devoted a section of the guideline on how to decide which agent to use; we also included
infliximab in our guideline as we feel this is also an important agent which is likely to
considered in managing the same patient group.
Additional trials have been published since the appraisal exercise was undertaken. These
provide new evidence on the longer term use of efalizumab and on the dose response to 25mg
twice weekly and 50mg twice weekly for etanercept.
The following late breaking publications were considered by our working group
Papp KA, Tyring SK, Lahfa M et al. A global phase III randomised controlled trial of etanercept in
psoriasis: safety, efficacy, and effect of dose reduction. Br J Dermatol 2005; In press.
Leonardi CL, Papp KA, Gordon KB et al. Extended efalizumab therapy improves chronic plaque
psoriasis: results from a randomised phase III trial. J Am Acad Dermatol 2005; 52: 425-33.
Menter A, Gordon KB, Carey W et al. Efficacy and safety observed during 24 weeks of efalizumab
therapy in patients with moderate to severe plaque psoriasis. Arch Dermatol 2005; 141: 31-8.
Gottlieb AB, Hamilton TK, Caro I et al. Long term continuous efalizumab therapy maintains efficacy
and safety in patients with moderate to severe chronic plaque psoriasis: updated results from an
ongoing trial. In press 2005.
2
Ormerod, L. P. and Joint Tuberculosis Committee of the British Thoracic Society. Recommendations
for assessing risk and for managing M Tuberculosis infection and disease in adult patients due to start
anti-TNF-alpha treatment. 2005 In press
Kyle S, Chandler D, Griffiths C, Helliwell P, Lewis J, McInnes I, Oliver S, Symmons D, McHugh NJ.
2005. Guideline for anti-TNF treatment in psoriatic arthritis. Rheumatology. 44;390-397.
Hongbo Y, Thomas CL, Harrison MA et al. Translating the science of quality of life into practice:
What do Dermatology Life Quality Index scores mean? J Invest Dermatol 2005. In press
Finlay AY. Current severe psoriasis and the Rule of Tens. Br J Dermatol 2005;152:861-867
Eligibility = PASI score of >20
We would strongly question the evidence for patients requiring a PASI score of >20 to be
eligible for treatment. Work by Professor Finlay 1 has demonstrated that patients with a PASI
score of greater than 10 are likely to be suffering severe impairment of quality of life and to
be considered for drugs such as methotrexate and ciclosporin or for hospital admission. The
latter was an important cost consideration in the economic assessment.
Clinical trials on which the evidence for efficacy of these agents were based had a lower
inclusion criterion e.g. PASI of >10 or >12. In section 4.2.3.2 the ICERs are based on
patients with PASI> 10 DLQI>15. The advice therefore seems to be divergent from the
evidence put forward for this assessment. Feldman suggested PASI > 11 or >10% Body
Surface Area (BSA) as indicating severe disease (Journal of Investigative Dermatology
1996;106:183-186). A BSA > 10% has also been defined as having severe disease in the
National Psoriasis Foundation. Facts on Psoriasis: Disease severity (2004).
In section 4.2.3.2 (Page 15) PASI > 10 is referred to as having ‘relatively severe psoriasis’ –
this is contradicted if you then define ‘severe’ as >20.
The committee drawing up guidelines on biological interventions in psoriasis took a PASI of
>10 as the threshold severity criterion for considering these new treatments. We did however,
emphasise that many of these patients will already be receiving systemic therapy, albeit with
poor disease control or significant adverse effects. To stop treatment in order to achieve the
suggested PASI threshold of >20 would risk acute severe disease relapse with a consequent
need for urgent hospitalisation. Rigid application of the >20 rule would disadvantage and
possibly endanger such patients and risk increasing their overall management costs.
We question why in section 1.1 the QoL measure is not used as an additional measure of
severity. The correlation of PASI with impact on quality of life is poor and it is as important
to consider how psoriasis affects the individual psychosocially as it is merely to consider
extent of disease Some patients will be severely disadvantaged and denied treatment if the
entrance requirement is set at this level and no allowance for other factors is considered. On
the one hand, we would not want our guideline to imply that all otherwise eligible patients
with PASI greater than 10 should be offered or would want biological interventions.
1
(Finlay AY. Current severe psoriasis and the Rule of Tens. Br J Dermatol 2005;152:861-867 and Hongbo Y, Thomas C L.
Harrison M A, Salek M S, Finlay A Y. Translating the science of quality of life into practice: What do Dermatology Life
Quality Index (DLQI) scores mean? JID 2005 (In press – available on request from Professor Finlay, Cardiff)
3
However, highly visible (face, hands) or symptomatic psoriasis (hands, feet, flexures and
genitalia) may have an impact which is poorly reflected in the PASI score and for which
quality of life scores offer a much better surrogate. Involvement of face and hands can
profoundly affect the patient emotionally, functionally and economically. We feel strongly
that the eligibility criteria should include an additional quality of life component. This would
make the screening more rigorous and has been adopted in the BSR guidelines for
rheumatoid disease.
The evidence base for suggesting a PASI 75 as a required response to therapy
As it is now well established that maximal improvement from both drugs may not be
achieved by 12 weeks, the stipulation that PASI-75 must be achieved by 12 weeks is
potentially wasting resources by forcing a treatment which will later be successful, to be
abandoned inappropriately. Longer term data in the recently reported studies (above) confirm
this. Further, in section 4.1.1.1 (page 9) and 4.1.1.4 (page 10) some studies extend beyond 12
weeks eg 12-24 weeks – Leonardi 2005, Lebwohl 2003. We feel a PASI- improvement of
50% by 12 weeks would be more sensible. These patients are already, by definition, hard-totreat patients. Improvement in PASI may fall short of 75% but the benefits on quality of life
may still be significant and not achievable with other agents. We agree that patients who fail
to achieve a reduction in PASI of 50% should discontinue treatment at the 12 week stage.
The advice you give to use PASI 75 was taken from the rheumatologists’ guidelines for joint
disease. These were written a few years ago for a different patient group. Our guidelines are
more recent and considered the effects of these drugs on the skin rather than the joints and
were formulated with the help of rheumatologists who were involved in producing the BSR
guidelines. We considered the published evidence that a PASI improvement of 50% was seen
by patients and physicians as a worthwhile response for patients with severe disease (Carlin
CS, Feldman SR, Krueger JG et al. A 50% reduction in the Psoriasis Area and Severity Index
(PASI 50) is a clinically significant endpoint in the assessment of psoriasis. J. Am. Acad.
Dermatol. 2004; 859-66).
Our suggested response criteria include the additional requirement that treatment should also
have improved quality of life. Although quality of life issues are given emphasis in the
Appraisal Committee’s pre-amble, they are not taken into account in defining either
eligibility or response.
We have also suggested that a PASI improvement of 50% be used prospectively to define
failure of other agents. If the response criterion recommended by the Appraisal Committee
were applied to standard drugs such as methotrexate and ciclosporin, then the number of
patients eligible for biologicals might paradoxically increase.
4
Etanercept is restricted to use at 25mg twice weekly
New data in the Papp study (above) shows a clear advantage of the 50mg dose of etanercept
and we feel this should be available to for those patients who only respond suboptimally to
the 25mg dose - particularly as your current recommendation would class these as treatment
failures and may lead onto switching agents, repeating screening, risking a flaring of the
psoriasis and further time taken to assess whether it is going to work. We feel it appropriate
to restrict the higher dose but not to put it out of bounds.
In section 1.1 Etanercept is also licensed for use at 50mg twice per week and there may be
individual circumstances where the higher dose would be helpful. The recommendation of
‘not exceeding 25 mg twice weekly’ seems too proscriptive.
2 Whether you consider that the summaries of clinical and cost effectiveness
are reasonable interpretations of the evidence and that the preliminary views
on the resource impact and implications for the NHS are appropriate.
The order in which treatments should be used
We question the evidence that underpins the recommendation that etanercept must be used
before efalizumab - there is more data coming into the public domain showing longer term
efficacy and safety of efalizumab. For example Menter (above) showing PASI 75%
improvement in 26% at 12 weeks and in 44% at 24 weeks with efalizumab, Leonardi (above)
showing PASI 75% improvement in 39% at 12 weeks. In the absence of any scientific
evidence base for differences in efficacy or safety, clinical judgement should direct the order
in which therapies be used. Was this decision based on cost models as there are no
comparative trials? It appears to have been based on the assessment group’s economic
modelling for intermittent treatment, which can only apply to etanercept. It is not clear
whether the ICER model took account of the less constant control that would occur with
intermittent treatment. It is reasonable to assume that among high need patients the disease
would undergo repeated exacerbations and these would each take 12-24 weeks to respond - a
less costly but also less effective scenario over 1 year. Sensitivity analysis of the economic
models should factor in the possibility that intermittent treatment over 12 months might
produce only 50% of the qualy that continuous treatment might.
In 4.3.5 you state “ efalizumab and etanercept continuous were dominated by intermittent
etanercept”. We are not convinced that it is fair to compare continuous and intermittent
treatment. A cost comparison of continuous efalizumab against continuous etanercept needs
to be quoted but has been missed in this evaluation. Likewise the ICERS for efalizumab in
patients with severe disease at risk of admission are not given in section 4.2.4.2, so it is not
possible for the reader to evaluate any basis for the economic conclusions.
Section 4.3.5 gives the criteria for efalizumab as “failed to respond to or … intolerant of
etanercept” while section 7.3.2 has “or has a contra-indication to etanercept”. There is a
difference between being intolerant and having a contra-indication. These sections should
be consistent, if etanercept were to remain the favoured agent after the considerations above.
5
If this is the first instance in which NICE has prescribed the order in which drugs have to be
used, there needs to be a better evidence base than that presented in this assessment.
3 Whether you consider that the provisional recommendations of the
Appraisal Committee are sound and constitute a suitable basis for the
preparation of guidance to the NHS.
There is much to commend in the appraisal particularly that the importance of psoriasis as a
clinical need is acknowledged and the need for both treatments to be recognised, as they will
benefit different patients and have distinct modes of action. We appreciate the significant
costs of these treatments and a need to restrict their use to the most needy patients and also
the need for pharmacovigilance. We are taking steps to set up a register which we hope will
be established not far behind the final appraisal from NICE. We recognise that the
consultation period should enhance the appraisal and with further consideration of
stakeholder comments this will be a suitable basis for guidance.
4
Additional comments on the evaluation report.
Major points
1. Eligibility
The wording in Section 1.a of Appendix C (audit criteria), if interpreted as
written, is extremely restrictive. As it stands a patient would have had to
have tried and failed “all other systemic therapy in routine use” or to be
intolerant or have a contraindication to “all other systemic therapy in routine
use”. As worded, a patient who had a contraindication to one systemic
therapy but had failed all the others would not qualify. The second “all” does
not appear in sections 1.1 and 1.2. The definition of standard therapy and of
its failure or contra-indication is covered on page 5 of our guideline with
greater clarity.
The drugs that must have been used require much clearer instructions. “All
other systemic therapy in routine use including ciclosporin, methotrexate and
PUVA” is confusing as there are many agents that are non-standard or not
licensed such as hydroxycarbamide and fumarates.
We worked hard to define eligibilty. Specifically, in our guideline, we define standard
treatment as acitretin, ciclosporin, methotrexate, narrow band UVB and psoralen
photochemotherapy (PUVA). We then stipulate that the patients should fulfil at least
one of the following clinical categories:
(i) have developed or are at higher than average risk of developing clinically
important drug-related toxicity and where alternative standard therapy cannot be used
6
(ii) are or have become intolerant to or cannot receive standard systemic therapy
(iii) are or have become unresponsive to standard therapy
(iv) have disease that is only controlled by repeated in-patient management
(v) have significant, coexistent, unrelated co-morbidity which precludes use of
systemic agents such as ciclosporin or methotrexate
(vi) have severe, unstable, life-threatening disease (erythrodermic or pustular
psoriasis)
(vii) have psoriatic arthritis fulfilling the British Society of Rheumatology (BSR)
eligibility criteria for treatment with anti-TNF11, in association with skin disease
We define being unresponsive to standard treatment as:
an unsatisfactory clinical response (a less than 50% improvement in baseline PASI
score or % body surface area where the PASI is not applicable, and 5 point
improvement in DLQI) to at least 3 months treatment in the therapeutic dose range to
the following treatments :
 ciclosporin: 2.5 mg/kg/day; up to 5 mg/kg/day
 methotrexate single weekly dose (PO, SC, IM) at 15 mg/wk; max 25–30 mg/wk
 acitretin: 25–50 mg daily
 narrow band UVB or psoralen photochemotherapy (non response, rapid relapse or
exceeding recommended maximum doses: 150-200 treatments for PUVA, 350
treatments for narrow band UVB.
We suggest that NICE consider adopting these better defined criteria or refer to our
guideline for the detailed criteria.
2. Some members feel that PASI scoring is too difficult to learn or time consuming. It is
not ideal as it is a non-linear scale, with area being difficult to accurately estimate
(large inter-observer variations occur).
However, the guideline development group felt that an objective score of efficacy was
essential to build into the registration process and to monitor outcomes in a
meaningful way and that PASI would enable results to be compared with trials, the
literature and between agents. Rheumatologists were similarly wary of DAS scores
for arthritis but this became established practice and nurses were trained to make the
relevant observations. Patients with Crohn’s disease similarly have validated severity
indices CDAI on which to base therapy. We feel that despite its known deficiencies
PASI scoring should be carried out and measured as part of patient monitoring. Our
guideline can be supported by paper or excel worksheets that facilitate the calculation
of PASI score and a program of nurse education.
7
3. Use in Children
Although not licensed in children there is concern that we will be able to continue to
use etanercept when appropriate in children with psoriasis. Etanercept is used by
rheumatologists for children with Juvenile RA and has been used at Great Ormond
Street in very occasional children with severe psoriasis unresponsive to methotrexate.
These cases are rare but we feel it would be reasonable to advise that (1) etanercept is
the favoured product at this stage as there is greater experience with it in children in
general and (2) the drug be initiated by special centres with experience of treatment of
children.
Alternatively it could be acknowledged that the use in children was out-with the scope
of this appraisal and that recommendations were not made perhaps making the case
for more clinical trials in children on which to base recommendations.
4. Economic considerations
In section 6. 2.12 (page 6) in assessing the costs of alternative therapies you may find
useful information from a poster presented at the EADV, Prague 2002 by Piercy et al.
‘Estimating the cost of moderate to severe chronic plaque psoriasis in the UK.’.
Briefly, annual cost of systemic treatment: acitretin £411,545; cyclosporin
£10,151,526; methotrexate £1,485,007; phototherapy (UVB/PUVA) £14,956,500;
inpatient £8,983,200.
5. Tuberculosis risk
We have worked with the British Thoracic Association who have guidelines
specific to reducing the risk of tuberculosis and developed an algorithm for the
stratification of risk which includes skin testing those who are not
immunosuppressed already and differential responses depending on prior
exposure to BCG vaccination. We obviously recommend that these
precautions be taken.
Minor points
1. In section 2.2 (page 3), drugs should be included as an aggravating cause of psoriasis
2. In section 7. 4.1.1.1 (page 9), doses of 1-2 mg/kg/wk used in studies by Leonardi
2005 and Lebwohl 2003. (There being no significant difference between the doses)
3. The remit of the Appraisal Committee is set out in the preamble to the Appraisal
Consultation Document (ACD): The Department of Health and the National
Assembly for Wales have asked the National Institute for Health and Clinical
Excellence (NICE or the Institute) to conduct an appraisal of efalizumab and
etanercept for the treatment of psoriasis and provide guidance on its use to the NHS in
8
England and Wales. This does not restrict it to plaque psoriasis and this seems to
have crept into the title, recommendations and audit criteria (Appendix C) without
justification in the ACD. Section 2 (Clinical Need and Practice), by contrast, refers
only to psoriasis, not to plaque psoriasis.
9